Pregnancy in women with immunoglobulin A nephropathy: are obstetrical complications associated with renal prognosis?

Pregnancy in women with immunoglobulin A nephropathy: are obstetrical complications associated... Abstract Background Recent studies regarding immunoglobulin A nephropathy (IgAN) suggest no relationship between pregnancy and disease progression, although complicated pregnancies and impaired renal function are closely related. Methods This study used a propensity-score-matched cohort analysis. Among biopsy-confirmed IgAN women in three hospitals in Korea, those who experienced pregnancy after their diagnosis were included in the study group. Renal outcome was the composite of serum creatinine doubling, estimated glomerular filtration rate (eGFR) halving and events of end-stage renal disease. Pregnancies with preterm birth, low birth weight and pre-eclampsia were defined as complicated. Results Overall, 59 IgAN women who became pregnant after their diagnosis, and the same number of IgAN women who did not experience pregnancy were included in the control group. Although pregnancy itself did not worsen renal outcomes [adjusted hazard ratio (HR): 1.51; 95% confidence interval (CI) 0.57–4.01; P = 0.41], mothers with complicated pregnancies experienced worse renal prognosis, even after adjustment for baseline and pre-gestational characteristics (adjusted HR: 5.07; 95% CI 1.81–14.22; P = 0.002). Moreover, this relationship was only significant in mothers with decreased renal function (eGFR <60 mL/min/1.73 m2) (adjusted HR: 18.70; 95% CI 1.63–214.40; P = 0.02), baseline hypertension (adjusted HR: 4.17; 95% CI 1.13–15.33; P = 0.03) and overt proteinuria (≥1 g/day) (adjusted HR: 4.21; 95% CI 1.24–14.27; P = 0.02). In contrast, patients who experienced pregnancies without complications showed better renal outcomes than did those without post-biopsy pregnancy (P = 0.01). Conclusion Obstetric complications in patients with high renal risk, rather than pregnancy itself, are associated with renal progression of IgAN women. IgA nephropathy, low birth weight, preeclampsia, pregnancy, preterm birth INTRODUCTION Immunoglobulin A nephropathy (IgAN) is the most common type of primary glomerulonephritis, and it is an important cause of kidney function deterioration, leading to end-stage renal disease (ESRD) [1, 2]. As most patients with IgAN are diagnosed during their 20s and 30s [2, 3], the common childbearing period in women, planning childbirth has been a crucial issue for women with IgAN. Previously, pregnancy in patients with chronic kidney disease (CKD) was thought to be associated with a failure of the kidneys to adapt to the physiological changes of pregnancy, consequently contributing to adverse renal outcomes [4, 5]. However, some studies have shown that pregnancy affected neither renal prognosis nor survival in patients with IgAN [6–10]. There are several limitations associated with these conflicting results, and these should be clarified and overcome in future studies to determine the complex effect of pregnancy on renal prognosis in women with IgAN. First, associations between complications during pregnancy and renal prognosis have not been well-described. Recent studies have shown that pregnancy in patients with CKD may lead to worse maternal–foetal outcomes even from the early stages [10–13]. Although a non-negligible number of complicated pregnancies were included in previous studies regarding IgAN, the impact of this gestational complication on renal prognosis was not emphasized [8, 9]. Next, such studies analysed data collected during a short follow-up period and for a small number of women with IgAN, most of whom had preserved renal function [6–9]. In the present study, we aimed to assess the relationship between pregnancy and renal prognosis in women with IgAN. We used propensity-score matching and stratified pregnancies into two subcategories (complicated and uncomplicated) to investigate further whether obstetric complications are associated with renal prognosis. MATERIALS AND METHODS Ethics statement This study was approved by the institutional review boards of the Seoul National University Hospital (SNUH, IRB No. H-1504-093-666), the Seoul National University Bundang Hospital (SNUBH, IRB No. B-1506/304-302) and the Asan Medical Center (AMC, IRB No. 2016-0959). This study was conducted in accordance with the principles of the Declaration of Helsinki. As we performed a retrospective study, informed consent for electronic health record (EHR) review was waived. In addition, phone interviews were performed after acquiring direct verbal consent from the subjects. Study population Data from all women with biopsy-proven IgAN from three hospitals were reviewed. IgAN was diagnosed by immunofluorescence microscopy showing mesangial IgA deposition as the predominant or co-dominant immunoglobulin. Patients aged <15 years at diagnosis, with evidence of a secondary cause of IgAN, with a follow-up duration of <6 months, with missing baseline information and those who were receiving renal replacement therapy at the time of diagnosis were excluded from the study. IgAN mothers who experienced pregnancy after their diagnoses were included in the study group, and the control group was formed by propensity-score-based matching. Data collection An EHR review was done to collect data from patients with IgAN. Demographic and clinical parameters were obtained at the time of kidney biopsy, including age, body mass index (BMI), diabetes mellitus, hypertension and medication history. Serum and urinary laboratory results were extensively reviewed from the time of renal biopsy to the time of the last follow-up. As methods for serum creatinine measurement changed over time, all serum creatinine levels were recalibrated for an isotope-dilution mass spectrometry assay (Roche Diagnostics). The estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease equation [14]. Proteinuria was assessed using either a random urine protein-to-creatinine ratio or a 24-h urinary protein measurement. Pathological parameters were reviewed as previously described [1, 15]. All native renal biopsies were processed using standard techniques for light microscopy, immunofluorescence and electron microscopy. Histopathological changes were evaluated by two pathologists, who reviewed the renal biopsy slides. Additionally, we collected pre-gestational clinical characteristics, including age, BMI, parity, degree of proteinuria and presence of hypertension. Pre-gestational renal function was evaluated by the last measured eGFR before the estimated conception date and within 3 years from the delivery date. When describing peri-gestational alterations in renal function, maximal eGFR values at baseline, at each trimester of pregnancy and after delivery were plotted. In addition, preterm birth was defined as a delivery with a gestational age <37 weeks [16, 17]; low birth weight was defined as a birth weight <2.5 kg [17, 18]; and pre-eclampsia was diagnosed by an obstetrician at the time of delivery by combining findings of hypertension, proteinuria and end-organ damage. A complicated pregnancy was defined as one in which the composition of the three above-mentioned adverse pregnancy outcomes occurred for a single pregnancy: preterm birth, low birth weight and pre-eclampsia. In addition, very low birth weight was defined as a birth weight <1.5 kg, and early preterm birth was defined as a delivery with gestational age <34 weeks. Small for gestational age was defined as a weight at delivery below the 10th percentile for the gestational age. Moreover, we performed phone interviews to obtain information regarding deliveries that occurred outside the hospitals. A total of five experienced clinical research nurses called the women with IgAN. The calls were attempted at least three times for each subject to reduce the rate of non-responders, and the survey proceeded only after obtaining informed consent from the subjects. Outcome assessment The following renal outcomes and their development data were evaluated: patients whose serum creatinine levels were doubled in relation to baseline levels; patients whose eGFR was halved in relation to baseline values; and patients who started maintenance renal replacement therapy. To include the outcomes associated with the progression of ESRD, we reviewed the EHR database and the national renal replacement treatment database maintained by the Korean Society of Nephrology [19]. Statistical analysis Categorical variables were shown as percentages and frequencies and were analysed using the chi-squared test. Continuous variables were presented as the mean ± standard deviation, or as median and quartile values, and were analysed using the Student’s t-test or the median test according to the normality of the data. Normality was assessed using the Shapiro–Wilk normality test. Propensity-score matching was performed with the ‘MatchIt’ package in R, using nearest-neighbour matching without replacement, and a calliper value of 0.2. To perform the survival analyses, we first generated Kaplan–Meier survival curves with the log-rank test. Next, we used a Cox proportional hazards regression analysis with the ‘survival’ package in R to evaluate the effects of pregnancy on renal outcomes. We treated pregnancy as a time-dependent covariate, and first events of pregnancy or complicated pregnancy were used to define the variable. In addition, to compare renal prognosis according to the presence of complicated pregnancies, we divided our matched cohort into three subgroups, as follows: women with IgAN who had at least one complicated pregnancy on record, who had only uncomplicated pregnancies on record and who had no pregnancy after biopsy-proven diagnosis. Multivariate analyses were performed with adjustment for characteristics known to be related to the progression of IgAN: age, BMI, history of hypertension, baseline eGFR and degree of proteinuria. As we used time-varying covariates in the study, in addition to the baseline characteristics, pre-gestational variables were also included in the adjustment when analysing the post-pregnancy renal survival [20]. In multivariate analyses, there were no missing values in the baseline information, and imputation by the last-observation-carried-forward method was used for missing information in the pre-gestational characteristics. Lastly, we tested the effect of pregnancy in subgroups defined by the presence of high-risk characteristics for disease progression: decreased eGFR (<60 mL/min/1.73 m2), baseline hypertension and overt proteinuria (≥1 g/day or 1 g/g). As there were mothers with multiple gestational events after their kidney biopsies, we additionally described the characteristics and outcomes of all pregnancy cases. The Kruskal–Wallis one-way analysis of variance and the chi-squared test for linear-by-linear associations were performed to compare pre-gestational characteristics and pregnancy outcomes according to the pre-gestational eGFR. A linear mixed model with fixed effects was generated using the restricted estimated maximal likelihood method to evaluate differences in eGFR across time phases, comparing complicated and uncomplicated pregnancies. Statistical analyses were performed using SPSS version 22 (IBM software, Armonk, NY, USA) or using R package version 3.2.5 (R Development Core Team, Vienna, Austria). For all analyses, a two-sided P-value with a statistically significant level of 0.05 was used. RESULTS Study population Figure 1 shows a flow diagram describing the selection of our study population. Among those who were not receiving renal replacement therapy at baseline, our study originally included a total of 802 women whose data were collected between 1979 and 2014 in the SNUH; 181 women whose data were collected between 2003 and 2015 in the SNUBH; and 513 women whose data were collected between 1998 and 2014 in the AMC who were diagnosed with IgAN with pathologic confirmation by percutaneous kidney biopsy at an age >15 years. After exclusion, 1112 women with IgAN were screened for events of pregnancy. Results of EHR review and phone interviews showed that 59 women with IgAN had confirmed pregnancy events after their IgAN diagnoses. Next, the control group was formed by propensity-score matching and ultimately included 59 IgAN women without post-biopsy gestation. FIGURE 1 View largeDownload slide Flow diagram describing the selection of the study population. FIGURE 1 View largeDownload slide Flow diagram describing the selection of the study population. Comparisons of baseline characteristics Before applying propensity-score matching (Supplementary Table S1), women who experienced pregnancy after IgAN diagnosis were younger than control group (P < 0.001). Further, they had lower BMIs (P = 0.03) and were diagnosed their IgAN at an earlier age (P = 0.002). Pathologic characteristics were not different between the two groups. After 1:1 propensity-score matching, these differences in baseline characteristics between the two groups were lost (Table 1). Table 1 Baseline clinical and pathological characteristics of the matched cohort according to the presence of pregnancy experience after confirmed diagnosis Variables  Matched cohorta   Pregnancy (−) (n = 59)  Pregnancy (+) (n = 59)  P  Clinical characteristics         Age (years)  26 (23–32)  28 (24–31)  0.71   BMI (kg/m2)  20.4 (19.3–22.7)  20.9 (19.7–23.3)  0.27   Year of biopsy (year)  2005 (2002–2008)  2005 (2003–2007)  >0.99   Creatinine (mg/dL)  0.80 (0.70–1.00)  0.90 (0.70–1.00)  0.71   eGFR (mL/min/1.73 m2)  85.0 (64.7–102.0)  80.0 (61.0–105.6)  0.46   CKD stage (n, %)      0.45    Stage 1  27 (45.8)  22 (37.3)      Stage 2  23 (39.0)  23 (39.0)      Stage 3 or more  9 (15.3)  14 (23.7)     Proteinuria (g/day)  0.87 (0.43–1.60)  1.09 (0.46–2.02)  0.14   RAS blockade (n, %)  17 (28.8)  13 (22.0)  0.40   Diabetes mellitus (n, %)  1 (1.7)  0 (0.0)  0.32   Hypertension (n, %)  33 (55.9)  36 (61.0)  0.58  Pathologic characteristics         Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.85   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.32   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.84  Variables  Matched cohorta   Pregnancy (−) (n = 59)  Pregnancy (+) (n = 59)  P  Clinical characteristics         Age (years)  26 (23–32)  28 (24–31)  0.71   BMI (kg/m2)  20.4 (19.3–22.7)  20.9 (19.7–23.3)  0.27   Year of biopsy (year)  2005 (2002–2008)  2005 (2003–2007)  >0.99   Creatinine (mg/dL)  0.80 (0.70–1.00)  0.90 (0.70–1.00)  0.71   eGFR (mL/min/1.73 m2)  85.0 (64.7–102.0)  80.0 (61.0–105.6)  0.46   CKD stage (n, %)      0.45    Stage 1  27 (45.8)  22 (37.3)      Stage 2  23 (39.0)  23 (39.0)      Stage 3 or more  9 (15.3)  14 (23.7)     Proteinuria (g/day)  0.87 (0.43–1.60)  1.09 (0.46–2.02)  0.14   RAS blockade (n, %)  17 (28.8)  13 (22.0)  0.40   Diabetes mellitus (n, %)  1 (1.7)  0 (0.0)  0.32   Hypertension (n, %)  33 (55.9)  36 (61.0)  0.58  Pathologic characteristics         Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.85   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.32   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.84  RAS, renin–angiotensin system. Values are presented as n (%) for categorical variables and as median (25–75% interquartile) for non-normally distributed continuous variables. a Propensity-score-based matching was done with following variables: age, BMI, history of hypertension, amount of proteinuria, serum creatinine and eGFR; baseline use of RAS blockade; and pathologic characteristics scoring (interstitial fibrosis, tubular atrophy and interstitial inflammation). Relationship between pregnancy and renal progression In the matched cohort, a total of 25 women with IgAN had adverse renal outcomes identified over a median follow-up duration of 7.2 (3.6–10.5) years. Overall renal outcomes were not significantly different between IgAN women with pregnancies and those without (P = 0.60, Figure 2). Also, a time-dependent Cox proportional hazards analysis with adjustment for baseline and pre-gestational characteristics showed that pregnancy events were not significantly associated with renal prognosis of the patients [adjusted hazard ratio (HR): 1.51; 95% confidence interval (CI) 0.57–4.01; P = 0.41]. FIGURE 2 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of post-biopsy pregnancy. The survival curve is given for IgAN women who experienced pregnancy after their diagnosis (black, thick continuous line) and for those without post-biopsy pregnancy (black, dotted line). FIGURE 2 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of post-biopsy pregnancy. The survival curve is given for IgAN women who experienced pregnancy after their diagnosis (black, thick continuous line) and for those without post-biopsy pregnancy (black, dotted line). Complicated pregnancy and renal prognosis Next, we tested whether a complicated pregnancy was related to renal outcomes in IgAN. IgAN women with gestational complications had more frequent hypertension histories before becoming pregnant than those without gestational complications. The other baseline and pre-gestational clinical characteristics did not significantly differ between the two groups (Table 2). Table 2 Baseline and pre-gestational characteristics of IgAN women and pregnancies according to the events of complicated pregnancy   With complicated pregnancies (n = 22)  With uncomplicated pregnancies (n = 37)  P  Baseline characteristicsa         Age (years)  28 (25–31)  27 (23–30)  0.71   BMI (kg/m2)  21.1 (20.1–23.4)  20.9 (19.5–23.4)  0.87   Year of biopsy (year)  2004 (2001–2006)  2005 (2003–2008)  0.65   Creatinine (mg/dL)  0.95 (0.70–1.30)  0.80 (0.70–0.96)  0.13   eGFR (mL/min/1.73 m2)  71.1 (46.5–202.6)  84.2 (71.5–107.8)  0.21   Proteinuria (g/day)  1.60 (0.90–2.64)  0.87 (0.36–1.63)  0.15   RAS blockade (n, %)  5 (22.7)  8 (21.6)  0.92   Hypertension (n, %)  15 (68.2)  21 (56.8)  0.38   Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.64   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.55   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.78  Pre-gestational characteristicsb         Age (years)  29 (26–32)  27 (23–30)  0.65   BMI (kg/m2)  21.9 (20.1–23.9)  22.4 (20.7–24.5)  0.79   Creatinine (mg/dL)  0.90 (0.70–1.17)  0.73 (0.63–0.84)  0.09   eGFR (mL/min/1.73 m2)  72.8 (53.1–99.6)  93.2 (77.4–110.8)  0.09   Proteinuria (g/day)  0.89 (0.38–1.91)  0.53 (0.22–1.43)  0.90   Hypertension (n, %)  18 (81.8)  12 (34.3)  <0.001    With complicated pregnancies (n = 22)  With uncomplicated pregnancies (n = 37)  P  Baseline characteristicsa         Age (years)  28 (25–31)  27 (23–30)  0.71   BMI (kg/m2)  21.1 (20.1–23.4)  20.9 (19.5–23.4)  0.87   Year of biopsy (year)  2004 (2001–2006)  2005 (2003–2008)  0.65   Creatinine (mg/dL)  0.95 (0.70–1.30)  0.80 (0.70–0.96)  0.13   eGFR (mL/min/1.73 m2)  71.1 (46.5–202.6)  84.2 (71.5–107.8)  0.21   Proteinuria (g/day)  1.60 (0.90–2.64)  0.87 (0.36–1.63)  0.15   RAS blockade (n, %)  5 (22.7)  8 (21.6)  0.92   Hypertension (n, %)  15 (68.2)  21 (56.8)  0.38   Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.64   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.55   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.78  Pre-gestational characteristicsb         Age (years)  29 (26–32)  27 (23–30)  0.65   BMI (kg/m2)  21.9 (20.1–23.9)  22.4 (20.7–24.5)  0.79   Creatinine (mg/dL)  0.90 (0.70–1.17)  0.73 (0.63–0.84)  0.09   eGFR (mL/min/1.73 m2)  72.8 (53.1–99.6)  93.2 (77.4–110.8)  0.09   Proteinuria (g/day)  0.89 (0.38–1.91)  0.53 (0.22–1.43)  0.90   Hypertension (n, %)  18 (81.8)  12 (34.3)  <0.001  RAS, renin–angiotensin system. Values are presented as n (%) for categorical variables and median (25–75% interquartile) for continuous variables. a Characteristics at the time of kidney biopsy when the diagnosis of IgA nephropathy was made. b Characteristics at the pre-gestational period (within 3 years from delivery date) before first complicated or uncomplicated pregnancy were described according to the group. In the Kaplan–Meier survival analysis (Figure 3), there was a significant difference in renal outcomes among the three subgroups (P < 0.001). For women without pregnancy experience, the renal survival rate was 80.3% at 10 years and 70.4% at 20 years. However, for mothers with complicated pregnancies, the renal survival rate was 55.3% at 10 years and 46.1% at 20 years. Interestingly, IgAN women who experienced pregnancy but no gestational complications had an excellent renal prognosis, even compared with those without post-biopsy pregnancy experience (P = 0.01). Only one woman who delivered without obstetric complications, at 1.4 years after IgAN diagnosis, had doubled serum creatinine and halved eGFR at a follow-up 3.7 years after the IgAN diagnosis. Moreover, IgAN women with complicated pregnancies had worse renal outcomes than those who did not experience any pregnancy (P = 0.04). FIGURE 3 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of complicated or uncomplicated pregnancy. The survival curve is given for IgAN women who had at least one complicated pregnancy (black, continuous line), those who had only uncomplicated pregnancies (gray, broken line) and for those without post-biopsy pregnancy (black, dotted line). FIGURE 3 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of complicated or uncomplicated pregnancy. The survival curve is given for IgAN women who had at least one complicated pregnancy (black, continuous line), those who had only uncomplicated pregnancies (gray, broken line) and for those without post-biopsy pregnancy (black, dotted line). In the multivariate analyses, using the first complicated gestation as a time-dependent variable, complicated pregnancy remained a major risk factor for a worse renal outcome (adjusted HR: 5.07; 95% CI 1.81–14.22; P = 0.002), even after adjustment for multiple characteristics (Table 3). Interestingly, the association between complicated pregnancy and renal progression was different according to the presence of clinical risk factors. The association between complicated pregnancy and disease progression remained significant in those with a decreased eGFR (<60 mL/min/1.73 m2; adjusted HR: 18.70; 95% CI 1.63–214.40; P = 0.02), pre-existing hypertension (adjusted HR: 4.17; 95% CI 1.13–15.33; P = 0.03) and overt proteinuria of at least 1 g/day (adjusted HR: 4.21; 95% CI 1.24–14.27; P = 0.02). In contrast, for those without underlying risk factors, gestational complications were not evidently related to composite renal outcome. Table 3 Variables associated with renal prognosis in the matched cohort   Adjusted HRa  95% CI  P  Age (years)  1.05  0.99–1.12  0.13  BMI (kg/m2)  0.91  0.78–1.06  0.24  Proteinuria (g/day)  1.66  1.31–2.11  <0.001  eGFR (mL/min/1.73 m2)  1.00  0.99–1.01  0.76  Hypertension  2.16  0.73–6.40  0.16  Complicated pregnancy  5.07  1.81–14.22  0.002  Effect of complicated pregnancy in following subgroupsb   eGFR <60 mL/min/1.73 m2  18.70  1.63–214.40  0.02   eGFR ≥60 mL/min/1.73 m2  1.89  0.40–8.91  0.42   Hypertension (+)  4.17  1.13–15.33  0.03   Hypertension (–)  2.08  0.03–153.30  0.74   Proteinuria ≥1 g/day  4.21  1.24–14.27  0.02   Proteinuria <1 g/day  8.12  0.41–159.95  0.17    Adjusted HRa  95% CI  P  Age (years)  1.05  0.99–1.12  0.13  BMI (kg/m2)  0.91  0.78–1.06  0.24  Proteinuria (g/day)  1.66  1.31–2.11  <0.001  eGFR (mL/min/1.73 m2)  1.00  0.99–1.01  0.76  Hypertension  2.16  0.73–6.40  0.16  Complicated pregnancy  5.07  1.81–14.22  0.002  Effect of complicated pregnancy in following subgroupsb   eGFR <60 mL/min/1.73 m2  18.70  1.63–214.40  0.02   eGFR ≥60 mL/min/1.73 m2  1.89  0.40–8.91  0.42   Hypertension (+)  4.17  1.13–15.33  0.03   Hypertension (–)  2.08  0.03–153.30  0.74   Proteinuria ≥1 g/day  4.21  1.24–14.27  0.02   Proteinuria <1 g/day  8.12  0.41–159.95  0.17  a Adjusted for following time-varying covariates both from baseline and the pre-gestational period; age, BMI, proteinuria, eGFR and hypertension. b Subgroups were defined with baseline characteristics. Effects of complicated pregnancies were evaluated in each subgroup. Pregnancy outcomes of women with IgAN Next, we separately analysed the characteristics of each pregnancy (not the mothers) to assess the relationship between gestational complications and renal function further. Among the study group, 13 mothers had multiple gestation events, and a total of 75 pregnancies were identified, with 64 pregnancy events having information available regarding pre-gestational renal function. The characteristics of those 64 gestations and their pregnancy outcomes are shown in Table 4. Pregnancy events with relatively preserved pre-gestational eGFR (≥90 mL/min/1.73 m2) were frequently associated with adverse pregnancy outcomes (n = 9, 29%), and those with lower pre-gestational eGFRs showed even worse prognoses. The relationship remained valid for each pregnancy outcome, although statistical significance was not shown regarding the risks for preeclampsia (P = 0.23). Table 4 Pre-gestational characteristics and outcomes in pregnancies of IgAN women according to their pre-gestational renal function   eGFR <60 (n = 12a)  60≤ eGFR <90 (n = 21a)  eGFR ≥90 (n = 31a)  P  Age (years)  33 (31–36)  33 (30–34)  31 (29–35)  0.37  BMI (kg/m2)  22.3 (19.2–28.5)  22.7 (20.5–24.1)  22.9 (21.0–26.4)  0.65  Multiparity (n, %)  5 (41.7)  10 (47.6)  15 (46.9)  0.80  Pre-gestational creatinine (mg/dL)  1.60 (1.23–1.86)  0.90 (0.82–0.92)  0.63 (0.60–0.69)  <0.001  Pre-gestational eGFR (mL/min/1.73 m2)  38.7 (31.6–49.7)  76.6 (71.0–82.5)  108.6 (99.2–117.2)  <0.001  Hypertension (n, %)  9 (75.0)  14 (66.7)  12 (40.0)  0.02  Composite pregnancy outcome (n, %)  8 (66.7)  7 (33.3)  9 (29.0)     Pre-eclampsia (n, %)  5 (41.7)  2 (9.5)  6 (18.8)  0.23   Low birth weight (n, %)  7 (58.3)  3 (15.8)  6 (20.7)  0.04    Very low birth weight (n, %)  3 (25.0)  1 (4.8)  2 (6.5)  0.12    Small for gestational age (n, %)  5 (41.7)  1 (4.8)  0 (0.0)  <0.001    Birth weight (kg)  2.09 (1.43–2.93)  3.15 (2.70–3.33)  3.17 (2.60–3.53)  0.002   Preterm birth (n, %)  8 (66.7)  6 (28.6)  7 (21.9)  0.01    Early preterm (n, %)  3 (25.0)  2 (9.5)  3 (9.7)  0.24    Gestational age (weeks)  36 (32–38)  39 (36–40)  39 (37–40)  0.02    eGFR <60 (n = 12a)  60≤ eGFR <90 (n = 21a)  eGFR ≥90 (n = 31a)  P  Age (years)  33 (31–36)  33 (30–34)  31 (29–35)  0.37  BMI (kg/m2)  22.3 (19.2–28.5)  22.7 (20.5–24.1)  22.9 (21.0–26.4)  0.65  Multiparity (n, %)  5 (41.7)  10 (47.6)  15 (46.9)  0.80  Pre-gestational creatinine (mg/dL)  1.60 (1.23–1.86)  0.90 (0.82–0.92)  0.63 (0.60–0.69)  <0.001  Pre-gestational eGFR (mL/min/1.73 m2)  38.7 (31.6–49.7)  76.6 (71.0–82.5)  108.6 (99.2–117.2)  <0.001  Hypertension (n, %)  9 (75.0)  14 (66.7)  12 (40.0)  0.02  Composite pregnancy outcome (n, %)  8 (66.7)  7 (33.3)  9 (29.0)     Pre-eclampsia (n, %)  5 (41.7)  2 (9.5)  6 (18.8)  0.23   Low birth weight (n, %)  7 (58.3)  3 (15.8)  6 (20.7)  0.04    Very low birth weight (n, %)  3 (25.0)  1 (4.8)  2 (6.5)  0.12    Small for gestational age (n, %)  5 (41.7)  1 (4.8)  0 (0.0)  <0.001    Birth weight (kg)  2.09 (1.43–2.93)  3.15 (2.70–3.33)  3.17 (2.60–3.53)  0.002   Preterm birth (n, %)  8 (66.7)  6 (28.6)  7 (21.9)  0.01    Early preterm (n, %)  3 (25.0)  2 (9.5)  3 (9.7)  0.24    Gestational age (weeks)  36 (32–38)  39 (36–40)  39 (37–40)  0.02  Values are presented as n (%) for categorical variables and as median (25–75% interquartile) for non-normally distributed continuous variables. Associations with categories of eGFR were estimated by Kruskal–Wallis one-way analysis of variance for continuous variables or linear-by-linear association by chi-squared test for categorical variables. a Number of pregnancies. Peri-gestational renal function according to gestational complications We plotted the eGFR changes during the peri-gestational period of all pregnancy events in Figure 4. There were no significant differences regarding the eGFR at baseline (P = 0.21) or during the pre-gestational period (P = 0.09) between pregnancy events with or without obstetrical complications. However, the trend toward gestational eGFR alteration was different between the two groups. There was a significant increase in eGFR from baseline until the third trimester in pregnancies without adverse pregnancy outcomes (P < 0.001), as is common in normal pregnancies. However, this eGFR increase was not observed in the complicated pregnancies (P = 0.57). Taken together, throughout the gestational period, mothers failed to show prominent hyperfiltration during the complicated pregnancies as much as during gestation without adverse pregnancy outcomes (P = 0.01). FIGURE 4 View largeDownload slide Change in eGFR during the peri-gestational periods of uncomplicated (gray, dotted line) and complicated pregnancies (black, continuous line). Vertical lines represent 95% CIs, and circles represent mean values. The numbers beside each circle represents the available number of pregnancies for which eGFR was measured at that particular phase of gestation. FIGURE 4 View largeDownload slide Change in eGFR during the peri-gestational periods of uncomplicated (gray, dotted line) and complicated pregnancies (black, continuous line). Vertical lines represent 95% CIs, and circles represent mean values. The numbers beside each circle represents the available number of pregnancies for which eGFR was measured at that particular phase of gestation. DISCUSSION In the present study, we demonstrated that pregnancy itself did not deteriorate renal function in women with IgAN, as reported in previous studies. Notably, we found that women with IgAN experienced many adverse obstetric complications during pregnancy. These complications manifested chiefly as a loss of eGFR elevation during pregnancies, implicating failure of renal adaptation to occur in response to pregnancy-related haemodynamic changes. Furthermore, these obstetric complications were associated with a worse renal prognosis in mothers with IgAN, especially in those with underlying risk factors, including established renal dysfunction, hypertension and overt proteinuria. Interestingly, however, women with IgAN with uncomplicated pregnancies showed even better renal prognosis than did those without pregnancy events after their diagnosis. Ours is the first study to suggest that the effect of pregnancy on renal outcomes in women with IgAN may differ according to the presence of an obstetric complication. There was evidence that women who experienced gestational complications had adverse renal and other cardio-/cerebrovascular disease outcomes [4, 5, 21, 22]. Additionally, one of the studies showed that pregnancies, regardless of the presence of pre-eclampsia, resulting in low birth weight and preterm birth were strongly related to ESRD progression [23]. However, regarding IgAN, previous studies reported that pregnancy had no significant effect on renal progression by studying mothers with relatively preserved renal function for a short-term follow-up period and not considering the effects of gestational complications [6–9, 13]. In the current study, we provided a differentiated data set that is more compatible with those for the previous large cohort studies, demonstrating the adverse long-term outcomes of gestational complications. Several mechanisms could be suggested for the worse renal prognosis observed in those with gestational complications. First, and fairly reasonable, pregnancy complications and renal progression share common risk factors [9, 12, 24]. As a large proportion of mothers with adverse pregnancy outcomes had baseline or pre-gestational hypertension or overt proteinuria, high-risk mothers were more likely to experience complicated pregnancies and subsequent disease progression. Secondly, as the relationship remained significant even after adjustment for multiple risk factors, women with obstetrical complications might have a decreased renal ‘reservoir’ that is not apparent in clinical parameters [25]. In the current study, we observed that the increase in eGFR during gestation, also known as physiological renal hyperfiltration during pregnancy [26], failed to occur in women with IgAN with obstetric complications. Additionally, another recent study reported that mothers with pre-eclampsia who progressed to ESRD were diagnosed with underlying kidney disease later in life [27]. Taken together, the impaired adaptation to the haemodynamic challenge during gestation might reflect subclinical renal dysfunction in those who failed to maintain safe pregnancy [25]. Lastly, complications during pregnancy might have directly damaged the kidney. As the acute postpartum deterioration of renal dysfunction in mothers with advanced CKD stages has been previously reported [4], the critical events accompanying acute kidney injuries could explain the association, although not all of them, in certain patients. Interestingly, we found that women with IgAN without obstetric complications showed excellent renal outcomes, even compared with the outcomes of women who did not become pregnant subsequent to the IgAN diagnosis. Considering that previous studies have shown that the rate of pregnancy complications is higher in women with CKD, even if the disease is in early stages [11, 12], the lack of pregnancy complication may, in itself, indicate that the mother has a relatively stable disease status. There were several limitations regarding our study. First, our study did not include a large number of women who became pregnant subsequent to a biopsy-confirmed IgAN diagnosis. This was mainly due to excluding patients for whom the follow-up was short, or for whom data were incomplete, which was necessary to ensure a robust comparison with propensity-score-based matching and to assess long-term renal prognosis. Secondly, while we reviewed all EHRs of the patients in our cohort, there was a non-negligible number, 504 (45.6%) patients in the total cohort, of patients who were not covered by the phone poll due to refusal or contact failure. However, most of them were older women and, among 619 patients who were aged <40 years, the number of missing young women was 163 (26.3%). Reducing this proportion would enable more valid studies with larger sample sizes. Thirdly, as our major findings were from a post hoc analysis, further study with a sufficient number of mothers with IgAN is necessary to provide confirmatory results. Fourth, the pregnancy events prior to the diagnoses were not evaluated, as the baseline characteristics were collected at the time of kidney biopsy. Lastly, the long time period and non-standardized treatment strategy due to the study’s retrospective nature should be considered. In conclusion, obstetric complications, which develop in >30% of pregnancies in women with IgAN, rather than pregnancy itself are associated with renal disease progression. This relationship is more prominent in mothers with IgAN with underlying risk factors. Therefore, clinicians should carefully monitor renal progression in women with IgAN who experienced obstetric complications. SUPPLEMENTARY DATA Supplementary data are available online at http://ndt.oxfordjournals.org. FUNDING This work was supported by the National Research Foundation of Korea, Republic of Korea (2014R1A1A1008201) and the Korean Health Technology R&D Project of the Ministry of Health and Welfare, Republic of Korea (HI15C2632). CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1 Lee H, Kim DK, Oh KH et al.   Mortality of IgA nephropathy patients: a single center experience over 30 years. PLoS One  2012; 7: e51225 Google Scholar CrossRef Search ADS PubMed  2 Nair R, Walker PD. Is IgA nephropathy the commonest primary glomerulopathy among young adults in the USA? Kidney Int  2006; 69: 1455– 1458 Google Scholar CrossRef Search ADS PubMed  3 Chang JH, Kim DK, Kim HW et al.   Changing prevalence of glomerular diseases in Korean adults: a review of 20 years of experience. Nephrol Dial Transplant  2009; 24: 2406– 2410 Google Scholar CrossRef Search ADS PubMed  4 Jones DC, Hayslett JP. Outcome of pregnancy in women with moderate or severe renal insufficiency. N Engl J Med  1996; 335: 226– 232 Google Scholar CrossRef Search ADS PubMed  5 Imbasciati E, Gregorini G, Cabiddu G et al.   Pregnancy in CKD stages 3 to 5: fetal and maternal outcomes. Am J Kidney Dis  2007; 49: 753– 762 Google Scholar CrossRef Search ADS PubMed  6 Abe S. Pregnancy in IgA nephropathy. Kidney Int  1991; 40: 1098– 1102 Google Scholar CrossRef Search ADS PubMed  7 Abe S. The influence of pregnancy on the long-term renal prognosis of IgA nephropathy. Clin Nephrol  1994; 41: 61– 64 Google Scholar PubMed  8 Limardo M, Imbasciati E, Ravani P et al.   Pregnancy and progression of IgA nephropathy: results of an Italian multicenter study. Am J Kidney Dis  2010; 56: 506– 512 Google Scholar CrossRef Search ADS PubMed  9 Liu Y, Ma X, Lv J et al.   Risk factors for pregnancy outcomes in patients with IgA nephropathy: a matched cohort study. Am J Kidney Dis  2014; 64: 730– 736 Google Scholar CrossRef Search ADS PubMed  10 Liu Y, Ma X, Zheng J et al.   A systematic review and meta-analysis of kidney and pregnancy outcomes in IgA nephropathy. Am J Nephrol  2016; 44: 187– 193 Google Scholar CrossRef Search ADS PubMed  11 Piccoli GB, Attini R, Vasario E et al.   Pregnancy and chronic kidney disease: a challenge in all CKD stages. Clin J Am Soc Nephrol  2010; 5: 844– 855 Google Scholar CrossRef Search ADS PubMed  12 Piccoli GB, Cabiddu G, Attini R et al.   Risk of adverse pregnancy outcomes in women with CKD. J Am Soc Nephrol  2015; 26: 2011– 2022 Google Scholar CrossRef Search ADS PubMed  13 Zhang JJ, Ma XX, Hao L et al.   A systematic review and meta-analysis of outcomes of pregnancy in CKD and CKD outcomes in pregnancy. Clin J Am Soc Nephrol  2015; 10: 1964– 1978 Google Scholar CrossRef Search ADS PubMed  14 Levey AS, Coresh J, Balk E et al.   National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med  2003; 139: 137– 147 Google Scholar CrossRef Search ADS PubMed  15 Lee H, Kim DK, Oh KH et al.   Mortality and renal outcome of primary glomerulonephritis in Korea: observation in 1,943 biopsied cases. Am J Nephrol  2013; 37: 74– 83 Google Scholar CrossRef Search ADS PubMed  16 Blencowe H, Cousens S, Oestergaard MZ et al.   National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet  2012; 379: 2162– 2172 Google Scholar CrossRef Search ADS PubMed  17 World Health Organization. International Statistical Classification of Diseases and Related Health Problems.  10th edn. Geneva. World Health Organization, 1992 18 Aylward GP, Pfeiffer SI, Wright A et al.   Outcome studies of low birth weight infants published in the last decade: a metaanalysis. J Pediatr  1989; 115: 515– 520 Google Scholar CrossRef Search ADS PubMed  19 Jin DC. Current status of dialysis therapy in Korea. Korean J Intern Med  2011; 26: 123– 131 Google Scholar CrossRef Search ADS PubMed  20 Dekker FW, de Mutsert R, van Dijk PC et al.   Survival analysis: time-dependent effects and time-varying risk factors. Kidney Int  2008; 74: 994– 997 Google Scholar CrossRef Search ADS PubMed  21 Smith GC, Pell JP, Walsh D. Pregnancy complications and maternal risk of ischaemic heart disease: a retrospective cohort study of 129,290 births. Lancet  2001; 357: 2002– 2006 Google Scholar CrossRef Search ADS PubMed  22 Pell JP, Smith GC, Walsh D. Pregnancy complications and subsequent maternal cerebrovascular events: a retrospective cohort study of 119,668 births. Am J Epidemiol  2004; 159: 336– 342 Google Scholar CrossRef Search ADS PubMed  23 Vikse BE, Irgens LM, Leivestad T et al.   Preeclampsia and the risk of end-stage renal disease. N Engl J Med  2008; 359: 800– 809 Google Scholar CrossRef Search ADS PubMed  24 Barbour SJ, Reich HN. Risk stratification of patients with IgA nephropathy. Am J Kidney Dis  2012; 59: 865– 873 Google Scholar CrossRef Search ADS PubMed  25 Ronco C, Brendolan A, Bragantini L et al.   Renal functional reserve in pregnancy. Nephrol Dial Transplant  1988; 3: 157– 161 Google Scholar CrossRef Search ADS PubMed  26 Odutayo A, Hladunewich M. Obstetric nephrology: renal hemodynamic and metabolic physiology in normal pregnancy. Clin J Am Soc Nephrol  2012; 7: 2073– 2080 Google Scholar CrossRef Search ADS PubMed  27 Kattah AG, Scantlebury DC, Agarwal S et al.   Preeclampsia and ESRD: the role of shared risk factors. Am J Kidney Dis  2016; doi:10.1053/j.ajkd.2016.07.034 © The Author 2017. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nephrology Dialysis Transplantation Oxford University Press

Pregnancy in women with immunoglobulin A nephropathy: are obstetrical complications associated with renal prognosis?

Loading next page...
 
/lp/ou_press/pregnancy-in-women-with-immunoglobulin-a-nephropathy-are-obstetrical-IWn4yPMoRv
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
ISSN
0931-0509
eISSN
1460-2385
D.O.I.
10.1093/ndt/gfx061
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Recent studies regarding immunoglobulin A nephropathy (IgAN) suggest no relationship between pregnancy and disease progression, although complicated pregnancies and impaired renal function are closely related. Methods This study used a propensity-score-matched cohort analysis. Among biopsy-confirmed IgAN women in three hospitals in Korea, those who experienced pregnancy after their diagnosis were included in the study group. Renal outcome was the composite of serum creatinine doubling, estimated glomerular filtration rate (eGFR) halving and events of end-stage renal disease. Pregnancies with preterm birth, low birth weight and pre-eclampsia were defined as complicated. Results Overall, 59 IgAN women who became pregnant after their diagnosis, and the same number of IgAN women who did not experience pregnancy were included in the control group. Although pregnancy itself did not worsen renal outcomes [adjusted hazard ratio (HR): 1.51; 95% confidence interval (CI) 0.57–4.01; P = 0.41], mothers with complicated pregnancies experienced worse renal prognosis, even after adjustment for baseline and pre-gestational characteristics (adjusted HR: 5.07; 95% CI 1.81–14.22; P = 0.002). Moreover, this relationship was only significant in mothers with decreased renal function (eGFR <60 mL/min/1.73 m2) (adjusted HR: 18.70; 95% CI 1.63–214.40; P = 0.02), baseline hypertension (adjusted HR: 4.17; 95% CI 1.13–15.33; P = 0.03) and overt proteinuria (≥1 g/day) (adjusted HR: 4.21; 95% CI 1.24–14.27; P = 0.02). In contrast, patients who experienced pregnancies without complications showed better renal outcomes than did those without post-biopsy pregnancy (P = 0.01). Conclusion Obstetric complications in patients with high renal risk, rather than pregnancy itself, are associated with renal progression of IgAN women. IgA nephropathy, low birth weight, preeclampsia, pregnancy, preterm birth INTRODUCTION Immunoglobulin A nephropathy (IgAN) is the most common type of primary glomerulonephritis, and it is an important cause of kidney function deterioration, leading to end-stage renal disease (ESRD) [1, 2]. As most patients with IgAN are diagnosed during their 20s and 30s [2, 3], the common childbearing period in women, planning childbirth has been a crucial issue for women with IgAN. Previously, pregnancy in patients with chronic kidney disease (CKD) was thought to be associated with a failure of the kidneys to adapt to the physiological changes of pregnancy, consequently contributing to adverse renal outcomes [4, 5]. However, some studies have shown that pregnancy affected neither renal prognosis nor survival in patients with IgAN [6–10]. There are several limitations associated with these conflicting results, and these should be clarified and overcome in future studies to determine the complex effect of pregnancy on renal prognosis in women with IgAN. First, associations between complications during pregnancy and renal prognosis have not been well-described. Recent studies have shown that pregnancy in patients with CKD may lead to worse maternal–foetal outcomes even from the early stages [10–13]. Although a non-negligible number of complicated pregnancies were included in previous studies regarding IgAN, the impact of this gestational complication on renal prognosis was not emphasized [8, 9]. Next, such studies analysed data collected during a short follow-up period and for a small number of women with IgAN, most of whom had preserved renal function [6–9]. In the present study, we aimed to assess the relationship between pregnancy and renal prognosis in women with IgAN. We used propensity-score matching and stratified pregnancies into two subcategories (complicated and uncomplicated) to investigate further whether obstetric complications are associated with renal prognosis. MATERIALS AND METHODS Ethics statement This study was approved by the institutional review boards of the Seoul National University Hospital (SNUH, IRB No. H-1504-093-666), the Seoul National University Bundang Hospital (SNUBH, IRB No. B-1506/304-302) and the Asan Medical Center (AMC, IRB No. 2016-0959). This study was conducted in accordance with the principles of the Declaration of Helsinki. As we performed a retrospective study, informed consent for electronic health record (EHR) review was waived. In addition, phone interviews were performed after acquiring direct verbal consent from the subjects. Study population Data from all women with biopsy-proven IgAN from three hospitals were reviewed. IgAN was diagnosed by immunofluorescence microscopy showing mesangial IgA deposition as the predominant or co-dominant immunoglobulin. Patients aged <15 years at diagnosis, with evidence of a secondary cause of IgAN, with a follow-up duration of <6 months, with missing baseline information and those who were receiving renal replacement therapy at the time of diagnosis were excluded from the study. IgAN mothers who experienced pregnancy after their diagnoses were included in the study group, and the control group was formed by propensity-score-based matching. Data collection An EHR review was done to collect data from patients with IgAN. Demographic and clinical parameters were obtained at the time of kidney biopsy, including age, body mass index (BMI), diabetes mellitus, hypertension and medication history. Serum and urinary laboratory results were extensively reviewed from the time of renal biopsy to the time of the last follow-up. As methods for serum creatinine measurement changed over time, all serum creatinine levels were recalibrated for an isotope-dilution mass spectrometry assay (Roche Diagnostics). The estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease equation [14]. Proteinuria was assessed using either a random urine protein-to-creatinine ratio or a 24-h urinary protein measurement. Pathological parameters were reviewed as previously described [1, 15]. All native renal biopsies were processed using standard techniques for light microscopy, immunofluorescence and electron microscopy. Histopathological changes were evaluated by two pathologists, who reviewed the renal biopsy slides. Additionally, we collected pre-gestational clinical characteristics, including age, BMI, parity, degree of proteinuria and presence of hypertension. Pre-gestational renal function was evaluated by the last measured eGFR before the estimated conception date and within 3 years from the delivery date. When describing peri-gestational alterations in renal function, maximal eGFR values at baseline, at each trimester of pregnancy and after delivery were plotted. In addition, preterm birth was defined as a delivery with a gestational age <37 weeks [16, 17]; low birth weight was defined as a birth weight <2.5 kg [17, 18]; and pre-eclampsia was diagnosed by an obstetrician at the time of delivery by combining findings of hypertension, proteinuria and end-organ damage. A complicated pregnancy was defined as one in which the composition of the three above-mentioned adverse pregnancy outcomes occurred for a single pregnancy: preterm birth, low birth weight and pre-eclampsia. In addition, very low birth weight was defined as a birth weight <1.5 kg, and early preterm birth was defined as a delivery with gestational age <34 weeks. Small for gestational age was defined as a weight at delivery below the 10th percentile for the gestational age. Moreover, we performed phone interviews to obtain information regarding deliveries that occurred outside the hospitals. A total of five experienced clinical research nurses called the women with IgAN. The calls were attempted at least three times for each subject to reduce the rate of non-responders, and the survey proceeded only after obtaining informed consent from the subjects. Outcome assessment The following renal outcomes and their development data were evaluated: patients whose serum creatinine levels were doubled in relation to baseline levels; patients whose eGFR was halved in relation to baseline values; and patients who started maintenance renal replacement therapy. To include the outcomes associated with the progression of ESRD, we reviewed the EHR database and the national renal replacement treatment database maintained by the Korean Society of Nephrology [19]. Statistical analysis Categorical variables were shown as percentages and frequencies and were analysed using the chi-squared test. Continuous variables were presented as the mean ± standard deviation, or as median and quartile values, and were analysed using the Student’s t-test or the median test according to the normality of the data. Normality was assessed using the Shapiro–Wilk normality test. Propensity-score matching was performed with the ‘MatchIt’ package in R, using nearest-neighbour matching without replacement, and a calliper value of 0.2. To perform the survival analyses, we first generated Kaplan–Meier survival curves with the log-rank test. Next, we used a Cox proportional hazards regression analysis with the ‘survival’ package in R to evaluate the effects of pregnancy on renal outcomes. We treated pregnancy as a time-dependent covariate, and first events of pregnancy or complicated pregnancy were used to define the variable. In addition, to compare renal prognosis according to the presence of complicated pregnancies, we divided our matched cohort into three subgroups, as follows: women with IgAN who had at least one complicated pregnancy on record, who had only uncomplicated pregnancies on record and who had no pregnancy after biopsy-proven diagnosis. Multivariate analyses were performed with adjustment for characteristics known to be related to the progression of IgAN: age, BMI, history of hypertension, baseline eGFR and degree of proteinuria. As we used time-varying covariates in the study, in addition to the baseline characteristics, pre-gestational variables were also included in the adjustment when analysing the post-pregnancy renal survival [20]. In multivariate analyses, there were no missing values in the baseline information, and imputation by the last-observation-carried-forward method was used for missing information in the pre-gestational characteristics. Lastly, we tested the effect of pregnancy in subgroups defined by the presence of high-risk characteristics for disease progression: decreased eGFR (<60 mL/min/1.73 m2), baseline hypertension and overt proteinuria (≥1 g/day or 1 g/g). As there were mothers with multiple gestational events after their kidney biopsies, we additionally described the characteristics and outcomes of all pregnancy cases. The Kruskal–Wallis one-way analysis of variance and the chi-squared test for linear-by-linear associations were performed to compare pre-gestational characteristics and pregnancy outcomes according to the pre-gestational eGFR. A linear mixed model with fixed effects was generated using the restricted estimated maximal likelihood method to evaluate differences in eGFR across time phases, comparing complicated and uncomplicated pregnancies. Statistical analyses were performed using SPSS version 22 (IBM software, Armonk, NY, USA) or using R package version 3.2.5 (R Development Core Team, Vienna, Austria). For all analyses, a two-sided P-value with a statistically significant level of 0.05 was used. RESULTS Study population Figure 1 shows a flow diagram describing the selection of our study population. Among those who were not receiving renal replacement therapy at baseline, our study originally included a total of 802 women whose data were collected between 1979 and 2014 in the SNUH; 181 women whose data were collected between 2003 and 2015 in the SNUBH; and 513 women whose data were collected between 1998 and 2014 in the AMC who were diagnosed with IgAN with pathologic confirmation by percutaneous kidney biopsy at an age >15 years. After exclusion, 1112 women with IgAN were screened for events of pregnancy. Results of EHR review and phone interviews showed that 59 women with IgAN had confirmed pregnancy events after their IgAN diagnoses. Next, the control group was formed by propensity-score matching and ultimately included 59 IgAN women without post-biopsy gestation. FIGURE 1 View largeDownload slide Flow diagram describing the selection of the study population. FIGURE 1 View largeDownload slide Flow diagram describing the selection of the study population. Comparisons of baseline characteristics Before applying propensity-score matching (Supplementary Table S1), women who experienced pregnancy after IgAN diagnosis were younger than control group (P < 0.001). Further, they had lower BMIs (P = 0.03) and were diagnosed their IgAN at an earlier age (P = 0.002). Pathologic characteristics were not different between the two groups. After 1:1 propensity-score matching, these differences in baseline characteristics between the two groups were lost (Table 1). Table 1 Baseline clinical and pathological characteristics of the matched cohort according to the presence of pregnancy experience after confirmed diagnosis Variables  Matched cohorta   Pregnancy (−) (n = 59)  Pregnancy (+) (n = 59)  P  Clinical characteristics         Age (years)  26 (23–32)  28 (24–31)  0.71   BMI (kg/m2)  20.4 (19.3–22.7)  20.9 (19.7–23.3)  0.27   Year of biopsy (year)  2005 (2002–2008)  2005 (2003–2007)  >0.99   Creatinine (mg/dL)  0.80 (0.70–1.00)  0.90 (0.70–1.00)  0.71   eGFR (mL/min/1.73 m2)  85.0 (64.7–102.0)  80.0 (61.0–105.6)  0.46   CKD stage (n, %)      0.45    Stage 1  27 (45.8)  22 (37.3)      Stage 2  23 (39.0)  23 (39.0)      Stage 3 or more  9 (15.3)  14 (23.7)     Proteinuria (g/day)  0.87 (0.43–1.60)  1.09 (0.46–2.02)  0.14   RAS blockade (n, %)  17 (28.8)  13 (22.0)  0.40   Diabetes mellitus (n, %)  1 (1.7)  0 (0.0)  0.32   Hypertension (n, %)  33 (55.9)  36 (61.0)  0.58  Pathologic characteristics         Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.85   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.32   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.84  Variables  Matched cohorta   Pregnancy (−) (n = 59)  Pregnancy (+) (n = 59)  P  Clinical characteristics         Age (years)  26 (23–32)  28 (24–31)  0.71   BMI (kg/m2)  20.4 (19.3–22.7)  20.9 (19.7–23.3)  0.27   Year of biopsy (year)  2005 (2002–2008)  2005 (2003–2007)  >0.99   Creatinine (mg/dL)  0.80 (0.70–1.00)  0.90 (0.70–1.00)  0.71   eGFR (mL/min/1.73 m2)  85.0 (64.7–102.0)  80.0 (61.0–105.6)  0.46   CKD stage (n, %)      0.45    Stage 1  27 (45.8)  22 (37.3)      Stage 2  23 (39.0)  23 (39.0)      Stage 3 or more  9 (15.3)  14 (23.7)     Proteinuria (g/day)  0.87 (0.43–1.60)  1.09 (0.46–2.02)  0.14   RAS blockade (n, %)  17 (28.8)  13 (22.0)  0.40   Diabetes mellitus (n, %)  1 (1.7)  0 (0.0)  0.32   Hypertension (n, %)  33 (55.9)  36 (61.0)  0.58  Pathologic characteristics         Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.85   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.32   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.84  RAS, renin–angiotensin system. Values are presented as n (%) for categorical variables and as median (25–75% interquartile) for non-normally distributed continuous variables. a Propensity-score-based matching was done with following variables: age, BMI, history of hypertension, amount of proteinuria, serum creatinine and eGFR; baseline use of RAS blockade; and pathologic characteristics scoring (interstitial fibrosis, tubular atrophy and interstitial inflammation). Relationship between pregnancy and renal progression In the matched cohort, a total of 25 women with IgAN had adverse renal outcomes identified over a median follow-up duration of 7.2 (3.6–10.5) years. Overall renal outcomes were not significantly different between IgAN women with pregnancies and those without (P = 0.60, Figure 2). Also, a time-dependent Cox proportional hazards analysis with adjustment for baseline and pre-gestational characteristics showed that pregnancy events were not significantly associated with renal prognosis of the patients [adjusted hazard ratio (HR): 1.51; 95% confidence interval (CI) 0.57–4.01; P = 0.41]. FIGURE 2 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of post-biopsy pregnancy. The survival curve is given for IgAN women who experienced pregnancy after their diagnosis (black, thick continuous line) and for those without post-biopsy pregnancy (black, dotted line). FIGURE 2 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of post-biopsy pregnancy. The survival curve is given for IgAN women who experienced pregnancy after their diagnosis (black, thick continuous line) and for those without post-biopsy pregnancy (black, dotted line). Complicated pregnancy and renal prognosis Next, we tested whether a complicated pregnancy was related to renal outcomes in IgAN. IgAN women with gestational complications had more frequent hypertension histories before becoming pregnant than those without gestational complications. The other baseline and pre-gestational clinical characteristics did not significantly differ between the two groups (Table 2). Table 2 Baseline and pre-gestational characteristics of IgAN women and pregnancies according to the events of complicated pregnancy   With complicated pregnancies (n = 22)  With uncomplicated pregnancies (n = 37)  P  Baseline characteristicsa         Age (years)  28 (25–31)  27 (23–30)  0.71   BMI (kg/m2)  21.1 (20.1–23.4)  20.9 (19.5–23.4)  0.87   Year of biopsy (year)  2004 (2001–2006)  2005 (2003–2008)  0.65   Creatinine (mg/dL)  0.95 (0.70–1.30)  0.80 (0.70–0.96)  0.13   eGFR (mL/min/1.73 m2)  71.1 (46.5–202.6)  84.2 (71.5–107.8)  0.21   Proteinuria (g/day)  1.60 (0.90–2.64)  0.87 (0.36–1.63)  0.15   RAS blockade (n, %)  5 (22.7)  8 (21.6)  0.92   Hypertension (n, %)  15 (68.2)  21 (56.8)  0.38   Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.64   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.55   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.78  Pre-gestational characteristicsb         Age (years)  29 (26–32)  27 (23–30)  0.65   BMI (kg/m2)  21.9 (20.1–23.9)  22.4 (20.7–24.5)  0.79   Creatinine (mg/dL)  0.90 (0.70–1.17)  0.73 (0.63–0.84)  0.09   eGFR (mL/min/1.73 m2)  72.8 (53.1–99.6)  93.2 (77.4–110.8)  0.09   Proteinuria (g/day)  0.89 (0.38–1.91)  0.53 (0.22–1.43)  0.90   Hypertension (n, %)  18 (81.8)  12 (34.3)  <0.001    With complicated pregnancies (n = 22)  With uncomplicated pregnancies (n = 37)  P  Baseline characteristicsa         Age (years)  28 (25–31)  27 (23–30)  0.71   BMI (kg/m2)  21.1 (20.1–23.4)  20.9 (19.5–23.4)  0.87   Year of biopsy (year)  2004 (2001–2006)  2005 (2003–2008)  0.65   Creatinine (mg/dL)  0.95 (0.70–1.30)  0.80 (0.70–0.96)  0.13   eGFR (mL/min/1.73 m2)  71.1 (46.5–202.6)  84.2 (71.5–107.8)  0.21   Proteinuria (g/day)  1.60 (0.90–2.64)  0.87 (0.36–1.63)  0.15   RAS blockade (n, %)  5 (22.7)  8 (21.6)  0.92   Hypertension (n, %)  15 (68.2)  21 (56.8)  0.38   Interstitial fibrosis (score)  1 (1–1)  1 (1–1)  0.64   Tubular atrophy (score)  1 (1–2)  1 (1–2)  0.55   Interstitial inflammation (score)  1 (1–2)  1 (1–2)  0.78  Pre-gestational characteristicsb         Age (years)  29 (26–32)  27 (23–30)  0.65   BMI (kg/m2)  21.9 (20.1–23.9)  22.4 (20.7–24.5)  0.79   Creatinine (mg/dL)  0.90 (0.70–1.17)  0.73 (0.63–0.84)  0.09   eGFR (mL/min/1.73 m2)  72.8 (53.1–99.6)  93.2 (77.4–110.8)  0.09   Proteinuria (g/day)  0.89 (0.38–1.91)  0.53 (0.22–1.43)  0.90   Hypertension (n, %)  18 (81.8)  12 (34.3)  <0.001  RAS, renin–angiotensin system. Values are presented as n (%) for categorical variables and median (25–75% interquartile) for continuous variables. a Characteristics at the time of kidney biopsy when the diagnosis of IgA nephropathy was made. b Characteristics at the pre-gestational period (within 3 years from delivery date) before first complicated or uncomplicated pregnancy were described according to the group. In the Kaplan–Meier survival analysis (Figure 3), there was a significant difference in renal outcomes among the three subgroups (P < 0.001). For women without pregnancy experience, the renal survival rate was 80.3% at 10 years and 70.4% at 20 years. However, for mothers with complicated pregnancies, the renal survival rate was 55.3% at 10 years and 46.1% at 20 years. Interestingly, IgAN women who experienced pregnancy but no gestational complications had an excellent renal prognosis, even compared with those without post-biopsy pregnancy experience (P = 0.01). Only one woman who delivered without obstetric complications, at 1.4 years after IgAN diagnosis, had doubled serum creatinine and halved eGFR at a follow-up 3.7 years after the IgAN diagnosis. Moreover, IgAN women with complicated pregnancies had worse renal outcomes than those who did not experience any pregnancy (P = 0.04). FIGURE 3 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of complicated or uncomplicated pregnancy. The survival curve is given for IgAN women who had at least one complicated pregnancy (black, continuous line), those who had only uncomplicated pregnancies (gray, broken line) and for those without post-biopsy pregnancy (black, dotted line). FIGURE 3 View largeDownload slide Kaplan–Meier survival curve of a matched cohort of women with IgAN according to the presence of complicated or uncomplicated pregnancy. The survival curve is given for IgAN women who had at least one complicated pregnancy (black, continuous line), those who had only uncomplicated pregnancies (gray, broken line) and for those without post-biopsy pregnancy (black, dotted line). In the multivariate analyses, using the first complicated gestation as a time-dependent variable, complicated pregnancy remained a major risk factor for a worse renal outcome (adjusted HR: 5.07; 95% CI 1.81–14.22; P = 0.002), even after adjustment for multiple characteristics (Table 3). Interestingly, the association between complicated pregnancy and renal progression was different according to the presence of clinical risk factors. The association between complicated pregnancy and disease progression remained significant in those with a decreased eGFR (<60 mL/min/1.73 m2; adjusted HR: 18.70; 95% CI 1.63–214.40; P = 0.02), pre-existing hypertension (adjusted HR: 4.17; 95% CI 1.13–15.33; P = 0.03) and overt proteinuria of at least 1 g/day (adjusted HR: 4.21; 95% CI 1.24–14.27; P = 0.02). In contrast, for those without underlying risk factors, gestational complications were not evidently related to composite renal outcome. Table 3 Variables associated with renal prognosis in the matched cohort   Adjusted HRa  95% CI  P  Age (years)  1.05  0.99–1.12  0.13  BMI (kg/m2)  0.91  0.78–1.06  0.24  Proteinuria (g/day)  1.66  1.31–2.11  <0.001  eGFR (mL/min/1.73 m2)  1.00  0.99–1.01  0.76  Hypertension  2.16  0.73–6.40  0.16  Complicated pregnancy  5.07  1.81–14.22  0.002  Effect of complicated pregnancy in following subgroupsb   eGFR <60 mL/min/1.73 m2  18.70  1.63–214.40  0.02   eGFR ≥60 mL/min/1.73 m2  1.89  0.40–8.91  0.42   Hypertension (+)  4.17  1.13–15.33  0.03   Hypertension (–)  2.08  0.03–153.30  0.74   Proteinuria ≥1 g/day  4.21  1.24–14.27  0.02   Proteinuria <1 g/day  8.12  0.41–159.95  0.17    Adjusted HRa  95% CI  P  Age (years)  1.05  0.99–1.12  0.13  BMI (kg/m2)  0.91  0.78–1.06  0.24  Proteinuria (g/day)  1.66  1.31–2.11  <0.001  eGFR (mL/min/1.73 m2)  1.00  0.99–1.01  0.76  Hypertension  2.16  0.73–6.40  0.16  Complicated pregnancy  5.07  1.81–14.22  0.002  Effect of complicated pregnancy in following subgroupsb   eGFR <60 mL/min/1.73 m2  18.70  1.63–214.40  0.02   eGFR ≥60 mL/min/1.73 m2  1.89  0.40–8.91  0.42   Hypertension (+)  4.17  1.13–15.33  0.03   Hypertension (–)  2.08  0.03–153.30  0.74   Proteinuria ≥1 g/day  4.21  1.24–14.27  0.02   Proteinuria <1 g/day  8.12  0.41–159.95  0.17  a Adjusted for following time-varying covariates both from baseline and the pre-gestational period; age, BMI, proteinuria, eGFR and hypertension. b Subgroups were defined with baseline characteristics. Effects of complicated pregnancies were evaluated in each subgroup. Pregnancy outcomes of women with IgAN Next, we separately analysed the characteristics of each pregnancy (not the mothers) to assess the relationship between gestational complications and renal function further. Among the study group, 13 mothers had multiple gestation events, and a total of 75 pregnancies were identified, with 64 pregnancy events having information available regarding pre-gestational renal function. The characteristics of those 64 gestations and their pregnancy outcomes are shown in Table 4. Pregnancy events with relatively preserved pre-gestational eGFR (≥90 mL/min/1.73 m2) were frequently associated with adverse pregnancy outcomes (n = 9, 29%), and those with lower pre-gestational eGFRs showed even worse prognoses. The relationship remained valid for each pregnancy outcome, although statistical significance was not shown regarding the risks for preeclampsia (P = 0.23). Table 4 Pre-gestational characteristics and outcomes in pregnancies of IgAN women according to their pre-gestational renal function   eGFR <60 (n = 12a)  60≤ eGFR <90 (n = 21a)  eGFR ≥90 (n = 31a)  P  Age (years)  33 (31–36)  33 (30–34)  31 (29–35)  0.37  BMI (kg/m2)  22.3 (19.2–28.5)  22.7 (20.5–24.1)  22.9 (21.0–26.4)  0.65  Multiparity (n, %)  5 (41.7)  10 (47.6)  15 (46.9)  0.80  Pre-gestational creatinine (mg/dL)  1.60 (1.23–1.86)  0.90 (0.82–0.92)  0.63 (0.60–0.69)  <0.001  Pre-gestational eGFR (mL/min/1.73 m2)  38.7 (31.6–49.7)  76.6 (71.0–82.5)  108.6 (99.2–117.2)  <0.001  Hypertension (n, %)  9 (75.0)  14 (66.7)  12 (40.0)  0.02  Composite pregnancy outcome (n, %)  8 (66.7)  7 (33.3)  9 (29.0)     Pre-eclampsia (n, %)  5 (41.7)  2 (9.5)  6 (18.8)  0.23   Low birth weight (n, %)  7 (58.3)  3 (15.8)  6 (20.7)  0.04    Very low birth weight (n, %)  3 (25.0)  1 (4.8)  2 (6.5)  0.12    Small for gestational age (n, %)  5 (41.7)  1 (4.8)  0 (0.0)  <0.001    Birth weight (kg)  2.09 (1.43–2.93)  3.15 (2.70–3.33)  3.17 (2.60–3.53)  0.002   Preterm birth (n, %)  8 (66.7)  6 (28.6)  7 (21.9)  0.01    Early preterm (n, %)  3 (25.0)  2 (9.5)  3 (9.7)  0.24    Gestational age (weeks)  36 (32–38)  39 (36–40)  39 (37–40)  0.02    eGFR <60 (n = 12a)  60≤ eGFR <90 (n = 21a)  eGFR ≥90 (n = 31a)  P  Age (years)  33 (31–36)  33 (30–34)  31 (29–35)  0.37  BMI (kg/m2)  22.3 (19.2–28.5)  22.7 (20.5–24.1)  22.9 (21.0–26.4)  0.65  Multiparity (n, %)  5 (41.7)  10 (47.6)  15 (46.9)  0.80  Pre-gestational creatinine (mg/dL)  1.60 (1.23–1.86)  0.90 (0.82–0.92)  0.63 (0.60–0.69)  <0.001  Pre-gestational eGFR (mL/min/1.73 m2)  38.7 (31.6–49.7)  76.6 (71.0–82.5)  108.6 (99.2–117.2)  <0.001  Hypertension (n, %)  9 (75.0)  14 (66.7)  12 (40.0)  0.02  Composite pregnancy outcome (n, %)  8 (66.7)  7 (33.3)  9 (29.0)     Pre-eclampsia (n, %)  5 (41.7)  2 (9.5)  6 (18.8)  0.23   Low birth weight (n, %)  7 (58.3)  3 (15.8)  6 (20.7)  0.04    Very low birth weight (n, %)  3 (25.0)  1 (4.8)  2 (6.5)  0.12    Small for gestational age (n, %)  5 (41.7)  1 (4.8)  0 (0.0)  <0.001    Birth weight (kg)  2.09 (1.43–2.93)  3.15 (2.70–3.33)  3.17 (2.60–3.53)  0.002   Preterm birth (n, %)  8 (66.7)  6 (28.6)  7 (21.9)  0.01    Early preterm (n, %)  3 (25.0)  2 (9.5)  3 (9.7)  0.24    Gestational age (weeks)  36 (32–38)  39 (36–40)  39 (37–40)  0.02  Values are presented as n (%) for categorical variables and as median (25–75% interquartile) for non-normally distributed continuous variables. Associations with categories of eGFR were estimated by Kruskal–Wallis one-way analysis of variance for continuous variables or linear-by-linear association by chi-squared test for categorical variables. a Number of pregnancies. Peri-gestational renal function according to gestational complications We plotted the eGFR changes during the peri-gestational period of all pregnancy events in Figure 4. There were no significant differences regarding the eGFR at baseline (P = 0.21) or during the pre-gestational period (P = 0.09) between pregnancy events with or without obstetrical complications. However, the trend toward gestational eGFR alteration was different between the two groups. There was a significant increase in eGFR from baseline until the third trimester in pregnancies without adverse pregnancy outcomes (P < 0.001), as is common in normal pregnancies. However, this eGFR increase was not observed in the complicated pregnancies (P = 0.57). Taken together, throughout the gestational period, mothers failed to show prominent hyperfiltration during the complicated pregnancies as much as during gestation without adverse pregnancy outcomes (P = 0.01). FIGURE 4 View largeDownload slide Change in eGFR during the peri-gestational periods of uncomplicated (gray, dotted line) and complicated pregnancies (black, continuous line). Vertical lines represent 95% CIs, and circles represent mean values. The numbers beside each circle represents the available number of pregnancies for which eGFR was measured at that particular phase of gestation. FIGURE 4 View largeDownload slide Change in eGFR during the peri-gestational periods of uncomplicated (gray, dotted line) and complicated pregnancies (black, continuous line). Vertical lines represent 95% CIs, and circles represent mean values. The numbers beside each circle represents the available number of pregnancies for which eGFR was measured at that particular phase of gestation. DISCUSSION In the present study, we demonstrated that pregnancy itself did not deteriorate renal function in women with IgAN, as reported in previous studies. Notably, we found that women with IgAN experienced many adverse obstetric complications during pregnancy. These complications manifested chiefly as a loss of eGFR elevation during pregnancies, implicating failure of renal adaptation to occur in response to pregnancy-related haemodynamic changes. Furthermore, these obstetric complications were associated with a worse renal prognosis in mothers with IgAN, especially in those with underlying risk factors, including established renal dysfunction, hypertension and overt proteinuria. Interestingly, however, women with IgAN with uncomplicated pregnancies showed even better renal prognosis than did those without pregnancy events after their diagnosis. Ours is the first study to suggest that the effect of pregnancy on renal outcomes in women with IgAN may differ according to the presence of an obstetric complication. There was evidence that women who experienced gestational complications had adverse renal and other cardio-/cerebrovascular disease outcomes [4, 5, 21, 22]. Additionally, one of the studies showed that pregnancies, regardless of the presence of pre-eclampsia, resulting in low birth weight and preterm birth were strongly related to ESRD progression [23]. However, regarding IgAN, previous studies reported that pregnancy had no significant effect on renal progression by studying mothers with relatively preserved renal function for a short-term follow-up period and not considering the effects of gestational complications [6–9, 13]. In the current study, we provided a differentiated data set that is more compatible with those for the previous large cohort studies, demonstrating the adverse long-term outcomes of gestational complications. Several mechanisms could be suggested for the worse renal prognosis observed in those with gestational complications. First, and fairly reasonable, pregnancy complications and renal progression share common risk factors [9, 12, 24]. As a large proportion of mothers with adverse pregnancy outcomes had baseline or pre-gestational hypertension or overt proteinuria, high-risk mothers were more likely to experience complicated pregnancies and subsequent disease progression. Secondly, as the relationship remained significant even after adjustment for multiple risk factors, women with obstetrical complications might have a decreased renal ‘reservoir’ that is not apparent in clinical parameters [25]. In the current study, we observed that the increase in eGFR during gestation, also known as physiological renal hyperfiltration during pregnancy [26], failed to occur in women with IgAN with obstetric complications. Additionally, another recent study reported that mothers with pre-eclampsia who progressed to ESRD were diagnosed with underlying kidney disease later in life [27]. Taken together, the impaired adaptation to the haemodynamic challenge during gestation might reflect subclinical renal dysfunction in those who failed to maintain safe pregnancy [25]. Lastly, complications during pregnancy might have directly damaged the kidney. As the acute postpartum deterioration of renal dysfunction in mothers with advanced CKD stages has been previously reported [4], the critical events accompanying acute kidney injuries could explain the association, although not all of them, in certain patients. Interestingly, we found that women with IgAN without obstetric complications showed excellent renal outcomes, even compared with the outcomes of women who did not become pregnant subsequent to the IgAN diagnosis. Considering that previous studies have shown that the rate of pregnancy complications is higher in women with CKD, even if the disease is in early stages [11, 12], the lack of pregnancy complication may, in itself, indicate that the mother has a relatively stable disease status. There were several limitations regarding our study. First, our study did not include a large number of women who became pregnant subsequent to a biopsy-confirmed IgAN diagnosis. This was mainly due to excluding patients for whom the follow-up was short, or for whom data were incomplete, which was necessary to ensure a robust comparison with propensity-score-based matching and to assess long-term renal prognosis. Secondly, while we reviewed all EHRs of the patients in our cohort, there was a non-negligible number, 504 (45.6%) patients in the total cohort, of patients who were not covered by the phone poll due to refusal or contact failure. However, most of them were older women and, among 619 patients who were aged <40 years, the number of missing young women was 163 (26.3%). Reducing this proportion would enable more valid studies with larger sample sizes. Thirdly, as our major findings were from a post hoc analysis, further study with a sufficient number of mothers with IgAN is necessary to provide confirmatory results. Fourth, the pregnancy events prior to the diagnoses were not evaluated, as the baseline characteristics were collected at the time of kidney biopsy. Lastly, the long time period and non-standardized treatment strategy due to the study’s retrospective nature should be considered. In conclusion, obstetric complications, which develop in >30% of pregnancies in women with IgAN, rather than pregnancy itself are associated with renal disease progression. This relationship is more prominent in mothers with IgAN with underlying risk factors. Therefore, clinicians should carefully monitor renal progression in women with IgAN who experienced obstetric complications. SUPPLEMENTARY DATA Supplementary data are available online at http://ndt.oxfordjournals.org. FUNDING This work was supported by the National Research Foundation of Korea, Republic of Korea (2014R1A1A1008201) and the Korean Health Technology R&D Project of the Ministry of Health and Welfare, Republic of Korea (HI15C2632). CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1 Lee H, Kim DK, Oh KH et al.   Mortality of IgA nephropathy patients: a single center experience over 30 years. PLoS One  2012; 7: e51225 Google Scholar CrossRef Search ADS PubMed  2 Nair R, Walker PD. Is IgA nephropathy the commonest primary glomerulopathy among young adults in the USA? Kidney Int  2006; 69: 1455– 1458 Google Scholar CrossRef Search ADS PubMed  3 Chang JH, Kim DK, Kim HW et al.   Changing prevalence of glomerular diseases in Korean adults: a review of 20 years of experience. Nephrol Dial Transplant  2009; 24: 2406– 2410 Google Scholar CrossRef Search ADS PubMed  4 Jones DC, Hayslett JP. Outcome of pregnancy in women with moderate or severe renal insufficiency. N Engl J Med  1996; 335: 226– 232 Google Scholar CrossRef Search ADS PubMed  5 Imbasciati E, Gregorini G, Cabiddu G et al.   Pregnancy in CKD stages 3 to 5: fetal and maternal outcomes. Am J Kidney Dis  2007; 49: 753– 762 Google Scholar CrossRef Search ADS PubMed  6 Abe S. Pregnancy in IgA nephropathy. Kidney Int  1991; 40: 1098– 1102 Google Scholar CrossRef Search ADS PubMed  7 Abe S. The influence of pregnancy on the long-term renal prognosis of IgA nephropathy. Clin Nephrol  1994; 41: 61– 64 Google Scholar PubMed  8 Limardo M, Imbasciati E, Ravani P et al.   Pregnancy and progression of IgA nephropathy: results of an Italian multicenter study. Am J Kidney Dis  2010; 56: 506– 512 Google Scholar CrossRef Search ADS PubMed  9 Liu Y, Ma X, Lv J et al.   Risk factors for pregnancy outcomes in patients with IgA nephropathy: a matched cohort study. Am J Kidney Dis  2014; 64: 730– 736 Google Scholar CrossRef Search ADS PubMed  10 Liu Y, Ma X, Zheng J et al.   A systematic review and meta-analysis of kidney and pregnancy outcomes in IgA nephropathy. Am J Nephrol  2016; 44: 187– 193 Google Scholar CrossRef Search ADS PubMed  11 Piccoli GB, Attini R, Vasario E et al.   Pregnancy and chronic kidney disease: a challenge in all CKD stages. Clin J Am Soc Nephrol  2010; 5: 844– 855 Google Scholar CrossRef Search ADS PubMed  12 Piccoli GB, Cabiddu G, Attini R et al.   Risk of adverse pregnancy outcomes in women with CKD. J Am Soc Nephrol  2015; 26: 2011– 2022 Google Scholar CrossRef Search ADS PubMed  13 Zhang JJ, Ma XX, Hao L et al.   A systematic review and meta-analysis of outcomes of pregnancy in CKD and CKD outcomes in pregnancy. Clin J Am Soc Nephrol  2015; 10: 1964– 1978 Google Scholar CrossRef Search ADS PubMed  14 Levey AS, Coresh J, Balk E et al.   National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med  2003; 139: 137– 147 Google Scholar CrossRef Search ADS PubMed  15 Lee H, Kim DK, Oh KH et al.   Mortality and renal outcome of primary glomerulonephritis in Korea: observation in 1,943 biopsied cases. Am J Nephrol  2013; 37: 74– 83 Google Scholar CrossRef Search ADS PubMed  16 Blencowe H, Cousens S, Oestergaard MZ et al.   National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet  2012; 379: 2162– 2172 Google Scholar CrossRef Search ADS PubMed  17 World Health Organization. International Statistical Classification of Diseases and Related Health Problems.  10th edn. Geneva. World Health Organization, 1992 18 Aylward GP, Pfeiffer SI, Wright A et al.   Outcome studies of low birth weight infants published in the last decade: a metaanalysis. J Pediatr  1989; 115: 515– 520 Google Scholar CrossRef Search ADS PubMed  19 Jin DC. Current status of dialysis therapy in Korea. Korean J Intern Med  2011; 26: 123– 131 Google Scholar CrossRef Search ADS PubMed  20 Dekker FW, de Mutsert R, van Dijk PC et al.   Survival analysis: time-dependent effects and time-varying risk factors. Kidney Int  2008; 74: 994– 997 Google Scholar CrossRef Search ADS PubMed  21 Smith GC, Pell JP, Walsh D. Pregnancy complications and maternal risk of ischaemic heart disease: a retrospective cohort study of 129,290 births. Lancet  2001; 357: 2002– 2006 Google Scholar CrossRef Search ADS PubMed  22 Pell JP, Smith GC, Walsh D. Pregnancy complications and subsequent maternal cerebrovascular events: a retrospective cohort study of 119,668 births. Am J Epidemiol  2004; 159: 336– 342 Google Scholar CrossRef Search ADS PubMed  23 Vikse BE, Irgens LM, Leivestad T et al.   Preeclampsia and the risk of end-stage renal disease. N Engl J Med  2008; 359: 800– 809 Google Scholar CrossRef Search ADS PubMed  24 Barbour SJ, Reich HN. Risk stratification of patients with IgA nephropathy. Am J Kidney Dis  2012; 59: 865– 873 Google Scholar CrossRef Search ADS PubMed  25 Ronco C, Brendolan A, Bragantini L et al.   Renal functional reserve in pregnancy. Nephrol Dial Transplant  1988; 3: 157– 161 Google Scholar CrossRef Search ADS PubMed  26 Odutayo A, Hladunewich M. Obstetric nephrology: renal hemodynamic and metabolic physiology in normal pregnancy. Clin J Am Soc Nephrol  2012; 7: 2073– 2080 Google Scholar CrossRef Search ADS PubMed  27 Kattah AG, Scantlebury DC, Agarwal S et al.   Preeclampsia and ESRD: the role of shared risk factors. Am J Kidney Dis  2016; doi:10.1053/j.ajkd.2016.07.034 © The Author 2017. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.

Journal

Nephrology Dialysis TransplantationOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial